probability panel
Ipsos
Ipsos What the Future: Gender (KnowledgePanel component, October 4-7, 2019) · Oct 4–7, 2019 · US national
Data flag. N AND MOE UNSOURCED. The cited Ipsos page contains no sample size and no margin of error; extensive searching found no document giving the recorded n=1,306 / +/-1.8 pairing. Treat both as unverified until a source is found.
Methodology
- Pollster
- Ipsos commercial
- Sponsor
- Ipsos
- Fielded
- Oct 4–7, 2019
- Population
- adults; includes an LGBT boost sample
- Sample size
- 1,306
- Sample frame
- probability panel
- Panel
- Ipsos KnowledgePanel
- Mode
- online - self-administered web survey of the probability-based KnowledgePanel; address-based sampling via the USPS Delivery Sequence File; tablet and internet provided to panelists without access
- Margin of error
- As published: 'plus or minus 1.8 percentage points at the 95% confidence level' for the general-population sample of adults, with a stated design effect of 1.12. NOTE: +/- 1.8 points is implausibly tight for n=1,021 (roughly +/- 3.2 would be expected at DEFF 1.12); the figure is recorded exactly as Ipsos published it and should be treated as a probable error in the source document.
- Weighting
- General population sample (N=1,021) weighted to adjust for gender by age, race-ethnicity, education, Census region, metropolitan status and household income, then raked to the geodemographic distributions of the 18+ adult population using the 2019 March Supplement of the Current Population Survey. LGBT boost (N=285) weighted to adjust for gender by age, race, education and Census region, then raked to the same distributions.
- Geography
- US national
- Base notes
- Nationally representative probability sample of adults 18+ (N=1,021) PLUS an LGBT boost sample (N=285), for a combined total of 1,306. IMPORTANT: the topline tables for the items below print '(N=1,306)', i.e. the combined base including the boost. Ipsos weighted the general-population sample and the LGBT boost separately, each raked to the 18+ adult population, but the release does not state how the two were combined for the published totals. Percentages on the contact and gender-belief items should therefore NOT be read as clean general-population estimates.
Sources
Verification note. Full 49-page Ipsos 'What the Future: Gender' topline retrieved and text-extracted. That single PDF documents FOUR separate Ipsos polls fielded Aug-Oct 2019 and released together on 2020-01-07; this record covers only the third of them, the KnowledgePanel probability poll of October 4-7, 2019, which carries the contact and gender-belief items. The male-only 'spectrum of genders' battery from the October 17-18 opt-in poll is recorded as a separate poll record. Two caveats matter: the published totals appear to include the LGBT boost sample, and the stated margin of error is internally implausible.
Figures below are transcribed from these documents. Where they disagree with the source, the source governs.
Questions asked (5)
knowing a transgender person, by relationship type
Do you know someone in your life who is...?
Do you know someone in your life who is...? Transgender
No
51%
Yes, an acquaintance
16%
Yes, a friend
9%
Yes, a coworker
5%
Yes, in my extended family
2%
Yes, in my immediate family
1%
Yes, a neighbor
1%
Not sure
12%
Refused
10%
Wording notes: response set distinguishes closeness of relationship rather than collapsing to a yes/no, which is a genuine design strength | includes 'Yes, I am' so self-identification is captured in the same item | 10% refused and 12% were not sure - high enough that the 'No' share is not a clean complement | published base appears to include the LGBT boost sample
neutral — Cleanly constructed contact item with graded relationship categories and no evaluative content. The striking substantive result is the steep gradient by identity: 14% say they do not know anyone gay, versus 51% who say they do not know anyone transgender - trans people are far less present in Americans' reported social networks than gay people, which is the mechanism most contact-hypothesis arguments turn on.
knowing a non-binary or gender non-conforming person
Do you know someone in your life who is...?
Do you know someone in your life who is...? Non-binary/Non-conforming
No
49%
Yes, an acquaintance
8%
Yes, a friend
4%
Yes, a coworker
2%
Yes, I am
1%
Yes, in my immediate family
1%
Yes, in my extended family
1%
Yes, a neighbor
1%
Not sure
27%
Refused
10%
Wording notes: 27% answered 'Not sure' - more than double the trans item's 12%, suggesting many respondents did not know what the category meant | published base appears to include the LGBT boost sample
neutral — Same neutral construction as the transgender item. The large 'not sure' share is itself the finding - in late 2019 a quarter of adults could not say whether they knew a non-binary person, which is a comprehension result as much as a contact result.
belief that gender is strictly binary
Please indicate the extent to which you agree or disagree with the following statements.
Please indicate the extent to which you agree or disagree with the following statements. - A person is either a man or a woman and not anything in between
Wording notes: offers a true midpoint ('Neither agree nor disagree') plus a separate 'Don't know', which most binary-belief items do not - 23% took one of the two | fielded alongside its near-converse ('There is a spectrum of gender identities'), which is a useful internal consistency check | published base appears to include the LGBT boost sample
neutral — Plainly stated proposition with a full symmetric scale, a midpoint and a don't-know. Its analytic value is that the same respondents were also asked the spectrum item: 41% agree gender is binary while 45% agree there is a spectrum of gender identities, which is only possible because a large bloc endorses both or neither. That overlap is direct evidence that these beliefs are unsettled rather than cleanly polarized.
belief that gender is defined by natal reproductive organs
Please indicate the extent to which you agree or disagree with the following statements.
Please indicate the extent to which you agree or disagree with the following statements. - Gender is defined by the reproductive organs you are born with
Wording notes: 'reproductive organs' is a narrower and more concrete criterion than 'sex assigned at birth', and narrower than chromosomes - respondents are answering about anatomy specifically | conflates sex and gender in the premise, which is precisely the proposition at issue | published base appears to include the LGBT boost sample
neutral — The statement asserts the essentialist position and asks for agreement, which is the standard way to measure it; the mirror position is captured by the adjacent spectrum item, so the battery as a whole is balanced. Worth noting that this anatomical framing draws higher agreement (50%) than the abstract binary framing (41%) among the same respondents - the concrete bodily criterion is more persuasive to people than the categorical claim.
belief that gender identity exists on a spectrum
Please indicate the extent to which you agree or disagree with the following statements.
Please indicate the extent to which you agree or disagree with the following statements. - There is a spectrum of gender identities
Wording notes: 10% chose 'Don't know', the highest of the battery - 'spectrum of gender identities' was still unfamiliar language in 2019 | the affirming counterpart to the binary and reproductive-organs items | published base appears to include the LGBT boost sample
neutral — Neutrally worded proposition with the same full scale as its counterparts. Recording all three items together is the point: the battery is one of the few instruments that measures essentialist and non-essentialist gender beliefs on identical scales within a single sample.